How well do adolescents recall use of mobile telephones? Results of a validation study

Abstract

Background: In the last decade mobile telephone use has become more widespread among children. Concerns expressed about possible health risks have led to epidemiological studies investigating adverse health outcomes associated with mobile telephone use. Most epidemiological studies have relied on self reported questionnaire responses to determine individual exposure. We sought to validate the accuracy of self reported adolescent mobile telephone use.

Methods: Participants were recruited from year 7 secondary school students in Melbourne, Australia. Adolescent recall of mobile telephone use was assessed using a self administered questionnaire which asked about number and average duration of calls per week. Validation of self reports was undertaken using Software Modified Phones (SMPs) which logged exposure details such as number and duration of calls.

Results: A total of 59 adolescents participated (39% boys, 61% girls). Overall a modest but significant rank correlation was found between self and validated number of voice calls (rho = 0.3, P = 0.04) with a sensitivity of 57% and specificity of 66%. Agreement between SMP measured and self reported duration of calls was poorer (rho = 0.1, P = 0.37). Participants whose parents belonged to the 4th socioeconomic stratum recalled mobile phone use better than others (rho = 0.6, P = 0.01).

Conclusion: Adolescent recall of mobile telephone use was only modestly accurate. Caution is warranted in interpreting results of epidemiological studies investigating health effects of mobile phone use in this age group.

Figure 1

Agreement between self reported and validated call indices per week: (a) number of calls (b) duration of calls.

Figure 2

Bland & Altman plot of ratio of self-reported to valid number of calls per week versus average of self-reported and valid number of calls per week.